Citation

Xiaoqiang Li, Jidong Wang, Mengjuan Li, Yang Jin, Zhijie Gu, Changchun Liu, Kenji Ogino. Fe-doped TiO₂/SiO₂ nanofibrous membranes with surface molecular imprinted modification for selective photodegradation of 4-nitrophenol[J]. Chin. Chem. Lett., 2018, 29(3): 527-530  

Xiaoqiang Li^a,b,1, Jidong Wang^b, Mengjuan Li^b, Yang Jin^b, Zhijie Gu^a, Changchun Liu^a, Kenji Ogino^a,1    

Waste-water from textile industry always contains pigments or dyes, which can cause severe environmental pollution; therefore, the water decontamination became one of the most urgent problems that need to be researched and dissolved. There are many efforts had been developed on dissolving the textile industry waste-water, such as using ultra-porous materials to adsorb dyes or degrading the dyes by electrochemical methods [1-5]. During recent decades, photocatalysis had been regarded as a green and sustainable technology and attracted much attention in environmental restoration [6-10].

TiO₂ is the most widely used semiconductor for photodegradation of organic pollutants, because of its non-toxicity, low cost, high photocatalytic activity and chemical stability. It is well known that only about 4% of the solar energy (in the UV region) can be utilized by TiO₂, because of its high intrinsic band gap of TiO₂ (3.2 eV for anatase, 3.02 eV for rutile) [11, 12]. In order to extent the absorbance range of TiO₂ to the visible region, various attempts have been employed including dye sensitization, doping with conductive polymer, non-metals, semiconductor compound, and transition metals. In our previous study, the polypyrrole was coated on the surface of TiO₂/SiO₂ (TS) nanofibrous membranes for the purpose of enhancing the overall photodegradation efficiency [13]. In recent years, the molecular imprinted polymers (MIPs) had been applied to enhance the photodegradation efficiency through improving the ability of recognition and the adsorption quantity for the aimed pollutants [14, 15]. MIPs usually synthesized or polymerized on the surface of photocatalysts with the target molecules [16].

In this work, o-phenylenediamine (OPDA) was chosen as the functional monomer to form MIPs layer on the surface of Fe-doped TiO₂/SiO₂ (Fe@TS) membrane. The -NH₂ groups of OPDA could interact with functional group of 4-nitrophenol(4NP) to form a precursor, ensuring the imprinting of the templates molecules during polymerization [17]. Furthermore, polymer with a polyaniline-like structure produced with the aid of UV light illumination of OPDA has the good chemical and photochemical stability. The processes of preparing molecular imprinted Fe@TS (MI-Fe@TS) membrane, were shown in Fig. 1. The chemicals and organics used in this work were purchased from Sinopharm Chemical Reagent Corporation and used without further purification.

Figure 1

	Download: JPG larger image
Fig. 1. Preparation processes of MI-Fe@TS

Due to the inherent properties of TiO₂, such as fragile and stiff, SiO₂ was applied to form the nanofibrous composite, which made the resultant photocatalysis have the excellent advantages of high surface-area-to-volume ratio and flexible enough for favorable recycling properties. In addition, Fe was used to extent the absorbent range of TiO₂. The electrospinning precursor solution was synthesized according to previous research [13]. In a typical process, 0.81 g of ferric nitrate nonahydrate (Fe(NO₃)₃·9H₂O) and 1.2 g polyvinylpyrrolidone (PVP, Mw = 1300000), were dissolved into 10 mL of N, N-dimethylformamide (DMF) with stirring to obtain solution A. In addition, 0.36 mL of H₂O with a pH of 2.3 (pH adjusted by HNO₃) was added to 1.89 g of silane coupling agent KH-560. Then, 0.42 g of tetraethoxysilane (Si(OC₂H₅)₄ TEOS) and 3.40 g of Titanium n-butoxide (Ti-(OC₄H₉)₄, TBOT), were added to the above solution and stirred to produce a homogeneous solution, to obtain solution B. After complete dissolution, solution A was mixed with solution B to obtain a homogeneous transparent solution, which was then incubated at 60 ℃ for 10 h to obtain the electrospinning precursor solution.

Thereafter, the solution was transferred into a plastic syringe for electrospinning at a fixed electrical voltage of 15 kV. The pump speed was 2.0 mL/h, and the distance between the needle tip and collector was kept at 20 cm. The fiber membrane was collected on aluminum foil and then peeled off, labeled as Fe@TSPVP. The membranes were then calcined at 800 ℃ for 2 h to remove PVP. The Fe@TS fibrous membranes were then obtained for further experiments. TiO₂/SiO₂ membranes were also prepared according the similar process, but without Fe doping, and labeled as TS.

The physicochemical properties of various membranes were then characterized. Morphology of nanofibers was observed through scanning electron microscopy (SEM, JSM-5900, Japan Electron Optics Laboratory Co., Ltd.). The SEM images of Fe@TS and MI-Fe@TS nanofibrous membranes are shown in Figs. 2a and b. Both of the nanofibrous samples present very smooth surface, but the diameter of Fe@TS nanofibers shrunk significantly after calcination. X-ray diffraction (XRD) patterns were recorded a Varian X-ray diffractometer (Rigaku D/Max 2200PC) with a graphite monochromator and Cu Ka radiation (λ = 0.15418 nm) in the range of 10°-80° at room temperature, and the voltage and electric current were fixed to 28 kV and 20 mA. As shown in Fig. 2b, Fe@TS sample exhibited anatase phase. According to the standard JCPSD Card No. 21-1271, characteristic diffraction peaks for anatase phase of TiO₂ are shown at 2θ = 25.2°, 37.8°, 47.9°, 54.0° and 62.6°, which corresponded to (101), (004), (200), (105) and (204), respectively. These results demonstrated that Fe dopant may inhibit the phase transformation of TiO₂ from anatase to rutile [18]. It was noted that Fe₂O₃ phase was not observed in the XRD pattern, which might be due to homogeneous dispersion of Fe³⁺ in TiO₂ lattice. The latter reasoning Energy dispersive X-ray spectroscopy (EDS) was conducted to analyze the chemical composition of Fe@TS and Fe@TSPVP nanofibers, which obviously proves that Fe@TS nanofibers are composed of Fe, Ti, Si and O. As shown in Fig. 2c, carbon (C) element was almost removed thoroughly by the calcination process. Based on the elemental analysis of the Fe@TS nanofibers, the mass fractions of Fe, Ti, Si, C and O were found to be 4.46, 31.13, 20.94, 2.02, and 41.44, respectively. Absence of other element peaks in thespectrum indicates the purity of the prepared nanocomposites. The oxidation state and concentration of the sample surface were observed using X-ray photoelectron spectroscopy (XPS, ESCALAB 250 multitechnique X-ray photoelectron spectrometer); the survey spectra were recorded with a pass energy of 160 eV, and the high-resolution spectra were recorded with a pass energy of 40 eV. The XPS spectra of Fe@TS are shown in Fig. 2d, the three different major peaks can be assigned to Fe 2p, Ti 2p, and Si 2p. Because of the calcination process, the C 1s peak of Fe@TS cannot be observed. The composition of the nanofibers was confirmed by their Fourier transform infrared (FTIR) spectra (Spectrum 100 FTIR spectrometer, PerkinElmer), as shown in Fig. 2e. The peaks at 3400, 2970 and 1630 cm^-1 belong to －OH, －C－H, and －C＝C－, disappeared after calcination, which demonstrates that PVP and KH560 molecules are removed by sintering. The peaks at 1110 and 1200 cm^-1 are due to asymmetric Si-O-Si stretching vibrations associated with the motion of oxygen in Si-O-Si antisymmetric stretching. The band at 799 cm^-1 is assigned to Si-O-Si symmetric stretching. The band at approximately 920 cm^-1 is assigned to the asymmetric Ti-O-Si vibration of TiO₂ and SiO₂ mixed oxides. The UV–vis spectra of TS and Fe@TS in the wavelength range of 200 nm to 800 nm were obtained from the dry-pressed disk samples using a UV–vis spectrophotometer (UV-3600) and shown in Fig. 2f. It can be detected that the Fe@TS nanofibers have a broader adsorption area than TS nanofibers, which indicates that the Fe@TS nanofibers allowed the excitation of the catalyst under visible light and thus, enhanced its photocatalytic activity.

Figure 2

	Download: JPG larger image
Fig. 2. a) SEM images of Fe@TSPVP, b) SEM image of Fe@TS, c) element analysis, d) XPS spectrum, e) FTIR spectra of Fe@TS and Fe@TSPVP membranes. f) UV–vis spectra of TS and Fe@TS nanofibers

The photodegradation activity of Fe@TS membranes was investigatedin a sealed quartz photoreactor filled with 30 mL of methyl orange solution (10 mg/L), and a 500 W xenon lamp was used as the light source. At given time intervals, the concentrations of methyl orange solutions were measured at 465 nm using the UV–vis absorption spectrophotometer. The TS nanofibers were used as comparison. As illustrated in Fig. 3a, there was not any significant change of methyl orange concentration, as the condition of absence of any catalyst under visible light, indicating that methyl orange is very stable and can remain in the natural environment. The photodegradation efficiency of methyl orange on TS membrane was about 8.6% with 90 min irradiation. The efficiency of the degradation of methyl orange with Fe@TS increased to about 34.1%. It should be noted that Fe₂O₃ has no photocatalytic activity. Thus, the enhanced degradation efficiency of Fe@TS is due to the Fe doping. The doping Fe can extend the visible absorption, depress hole-electron recombination, and narrow the band gap of TiO₂, as was confirmed with the UV–vis spectra of the Fe@TS nanofibers in Fig. 2f.

Figure 3

	Download: JPG larger image
Fig. 3. a) Photodegradation efficiency of methyl orange with Fe@TS, TS membranes and blank under visible light; b) Mechanism for the photocatalytic degradation of methyl orange on the Fe@TS nanofibers

The possible mechanism of methyl orange photodegradation on the Fe@TS nanofibers is proposed in Fig. 3b. First, water, O₂, and methyl orange are adsorbed on the surface of the Fe@TS nanofibers, followed by the formation of hole-electron pairs on nanofibers surface as the light energy to overcome the band gap energy between valence band and conduction band. Thereafter, the electrons excited from the nanofibers surface will transfer from the valence band into the conduction band. Then, the hydroxyl radicals with strong oxidizing activity further react with and degrade methyl orange into small molecules.

Thereafter, Fe@TS membranes were modified to form the layer of MIPs with 4NP as the template, through an in-situ polymerization process. In details, 0.24 g of OPDA and 0.04 g of 4NP were dissolved in 10 mL of distilled water, and the solution was stirred for 20 min, followed by adjusting the pH to 2.0 with HCl. Fe@TS membranes were further added into the solution, and then ultrasonicated for 30 min. After that, the polymerization was photocatalytically initiated by UV illumination with a 250 W Philips high-pressuremercury lamp. During imprinting procedure, a precursor was first formed through the hydrogen-bonding and electrostatic interactions between the molecules of OPDA and 4NP, and then the imprinted coating was obtained via the oxidative polymerization of OPDA. To remove template molecules in the composite, the resulting membranes were washed sequentially with ammonia solution, HCl solution, and distilled water, for three times. The obtained membranes were dried overnight at 70 ℃, and designated as MI-Fe@TS. For comparison purpose, the nonimprinted Fe@TS membrane (NI-Fe@TS) was also prepared with the similar processes, except without using the template of 4NP.

In this work, the photocatalytic ability was evaluated in single systems (only 4NP or methyl orange in the suspension) and in binary systems (both of 4NP and methyl orange in the suspension).

In the case of single system, photocatalytic experiments were conducted in a sealed quartz photoreactor filled with 30 mL of 4NP solution (10 mg/L), and a 500 W xenon lamp was used as the light source. Samples were immersed into the photoreactor in dark at 25 ℃ for 30 min to favor the organic adsorption onto the membranes surface. At given time intervals, the concentrations of 4NP solutions were measured at 315 nm using the UV–vis absorption spectrophotometer. The capability of various catalysts on photodegradation of methyl orange was also proceeded for the aim of comparison. Fig. 4a shows the kinetic data for photodegradation of 4NP over different photocatalysts (MI-Fe@TS, NIFe@TS and Fe@TS nanofibers) in single systems. The kinetic data showed that all the degradation processes followed the firstorder kinetics as equation:

(1)

Figure 4

	Download: JPG larger image
Fig. 4. Kinetic data for photocatalytic degradation of a) 4NP with different photocatalysts, b) methyl orange with different photocatalysts; c) UV–vis absorption spectrum in binary system with different photodegradation time

where k is the reaction rate constant, which can be used to measure the degradation rate. The k value of the target 4NP over nanofibrous photocatalysis are 0.00417, 0.00155 and 0.00263, for MI-Fe@TS, NI-Fe@TS and Fe@TS, respectively. Fig. 4b shows the kinetic data for the photodegradation of methyl orange by various nanofibrous photocatalysis in the single systems. The k value of methyl orange over MI-Fe@TS, NI-Fe@TS and Fe@TS were 0.0014, 0.00159 and 0.00265, respectively. The degradation of the target 4NP was remarkably enhanced by the MIPs layer while the degradation of the non-target methyl orange was correspondingly depressed, because of the effect of selective adsorption on the MIPs coatings.

In the binary system of competitive batch degradation tests, the solution of methyl orange and 4NP with the initial concentration of 10 mg/L was prepared; thereafter, MI-Fe@TS were added into the binary solution for photodegradation under xenon lamp. At given time intervals, the solution was analyzed using UV–vis spectrophotometer with the wavelength ranged from 250–550 nm. As shown in Fig. 4c, the 4NP degraded much quickly than methyl orange in the same system with MI-Fe@TS as the photocatalysis. This result also confirmed that MIPs have excellent photocatalytic selectivity towards the target contaminant.

The enhanced the degradation of target molecules and depressed the degradation of non-target dye is very likely because a large number of selective recognition sites existed in the surface of MI-Fe@TS nanofibers. The high sensitivity would benefit to removal of low-level target dyes in the presence of other high-level dyes. The whole photodegradation procedure includes two steps: First, the target compounds adsorbed by MI-Fe@TS fibers, and then be photodegraded by Fe-doped TiO₂. The adsorption process of targeted compounds onto the photocatalyst is the rate determining steps; therefore, the total degradation process would be greatly accelerated. For the MI-Fe@TS, many specific recognition sites existing on the nanofibers facilitated the adsorption of 4NP, which results in the acceleration of photodegradation of 4NP in the binary system.

In summary, 4NP imprinted Fe@TS nanofibers was prepared from the combined processes of sol-gel, electrospinning, calcination and liquid-phase deposition of molecular imprinted polymeric layer. Fe is doped into TS nanofibers to enhance their photodegradation capability. Furthermore, the MIPs layer was synthesized on the surface of Fe@TS nanofibers, which could molecules over the other organics. By simply changing template molecules, the TiO₂-based nanofibrous photocatalysts can be preparedfor other dyes or organics.

This work was supported by the National Natural Science Foundation of China (No. 51503083), China Postdoctoral Science Foundation (No. 2017M611696), the Fundamental Research Funds for the Central Universities (No. JUSRP51723B) and the National High-tech R&D Program of China (No. 2016YFB0302901).